From peaceful marches to violent clashes: a micro-situational analysis

Abstract

Recent studies point to the relevance of situational factors in the emergence of violence. This paper applies these insights to a systematic analysis of how and why peaceful protest marches transform into violent clashes. It focuses on the micro-situational patterns and emotional dynamics during protests. The exploratory study compares 20 peaceful and violent protests of the Global Justice Movement in the United States and Germany. It employs a triangulation of visual data with document data. The study relies on in-depth qualitative analysis, based on the principles of process tracing. Findings suggest that in those protests where violence emerges, a prior micro-situational pattern is systematically visible. The discussion of exemplary cases shows that two emotional phases precede the outbreak of violence. These phases emerge in a specific temporal danger zone of 1–3 h after the start of a protest. Further, specific triggering moments seem to prompt the outbreak of violence, like the breaking-up of police–protester lines, actors being outnumbered, or falling down. The emotional dynamic between protesters and officers during a protest likewise influences the intensity of violence and how violent situations end. Consequently, the paper suggests that actors need to go through a confrontational micro-situation in a demonstration in order to be able to use violence.

Keywords:

• Protest
• violence
• micro-sociology
• emotions
• visual data
• qualitative research

Acknowledgments:

The author would like to thank the anonymous reviewers for their helpful comments. The author is also thankful to Randall Collins for his invaluable advice on this study and paper and to Nicolas Legewie for his indispensable feedback.

Notes

1. Based on the scope conditions (left protest marches of the Global Justice Movement) I filtered over 100 protests with the media archive databases Lexis Nexis and Factiva. I then used the Microsoft Excel random sampling function in which each row in the list of over 100 cases had the same possibility of selection (see Nassauer, 2012 for more details).

2. The sample is not representative of the population of protests. I oversampled demonstrations with the focal outcome (i.e. violence) to systematically analyze how violence emerges in these cases.

3. Even behavioral observation software can nowadays code actors’ emotions in videos in a reliable manner (e.g. Noldus, 2014).

4. I will cite examples of the large amount of data throughout the paper. Due to copyright restrictions, video screenshots cannot be reprinted. The reader is encouraged to click on the YouTube links to see the data firsthand.

5. If violence is used from further away, more actors engage and it can be up to 10 people that use violence, e.g. by throwing stones at police.

6. The picture can be found online, among other places under: http://www.ludgerusschule.de/content/projekte/50jahre/80er/1981.htm.

7. For the very specific temporal sequence of a forward panic, see Nassauer, 2015b.